In October the White House released its report on AI: “Preparing for the Future of Artificial Intelligence.” The report had several goals: “to foster inter-agency coordination, to provide technical and policy advice on topics related to AI, and to monitor the development of AI technologies across Industry, the research community, and the Federal Government.
Here are the major takeaways and key points.
1. AI Should Be Used For The Public Good
Both public and private entity industries and businesses should benefit from AI. Plans to ensure access by creating an open – source AI training database are also in the works. Look for AI working for the public good in industries like criminal justice, the economy, healthcare, and transportation.
2. Government Should Embrace AI
AI makes everything faster and more efficient, so government agencies should embrace it. DARPA already has a digital tutor for Navy recruits, and the report recommends that tutor be adapted for every agency. Furthermore, the government will provide more long-term federal support for AI research even though the private sector will be the main source of funding.
3. We Must Regulate Automated And Unmanned Vehicles
Both to protect the public and ensure that competition is fair, we need new regulation in this area. Existing safety standards for cars should be updated to account for automated vehicles; and there is also a need for an automated and advanced air traffic management system. Putting the right regulations in place will mean developing a diverse federal workforce made up of senior people in the industry who can shape those new laws fairly.
4. Empowering The Next Generation
The report recommends that all American students, from kindergarten through High School, “learn computer science and be equipped with the computational thinking skills they need in a technology-driven world.” The flip side of the goal is to teach AI ethically. The reporter recommends that AI curricular including the study of computer science, data science, and machine learning at all schools and universities include a study of technology – focused ethics, privacy, safety, and security..
5. Use AI To Supplement Human Workers, Not Replace Them
One of the biggest fears Americans have about AI (aside from the whole Skynet thing) is that they will lose their jobs to AI workers. The report acknowledges that, in general, AI threatens low-wage jobs the most and might increase the wage gap.
However, evidence in the report and elsewhere clearly shows that AI is most effective when it is working in tandem with human intelligence. In fact, humans collaborating with AI provide the best results in many cases.
6. Data Should Be Free From Bias
AI systems are trained in closed systems, but then moved into our open world with its imperfect information. This opens up a dire need for accountability and fairness. AI should only use data that is complete and unbiased; poor quality data can make problems worse, causing AI to make bad decisions. Therefore, the report recommends that AI should use only data from which all bias has been eliminated – or it shouldn’t be used by AI at all.
7. Emphasis On Global Safety
Secrecy and intellectual property are important in the AI industry – but policy concerns means that confounds the monitoring for ethical standards. The report states that trade secrets will be allowed, but that clear industry milestone should be set. This way Watchdog entities can monitor progress without access to specific secrets. The report also sets forth a plan for international engagement and partnership with other countries active in AI.
The report is optimistic about AI, suggesting some caution, but acknowledging the fact that AI is here to stay and a benefit to humankind. The recommendation surrounding accountability and ethics are mostly common sense, and will not hurt the industry’s progress.